Evaluation of the Weather Research and Forecasting Model over the Contingent United Sates
Location
Eccles Conference Center
Event Website
http://water.usu.edu/
Start Date
3-29-2011 10:35 AM
End Date
3-29-2011 10:40 AM
Description
This study is intended to examine the performance of the next-generation Weather Research and Forecasting (WRF) model version 3.2 over the Contingent United States (CONUS). WRF was developed by the National Center for Atmospheric Research and has been broadly used for weather, climate, and water resources research and prediction. WRF's ability to combine user specified combinations of various physics and model settings is the main reason it is the preferred model by climate scientists, as it allows for a level of customization. Numerous permutations of the physics options within WRF were tested for the period of 1998 - 1999.
Through these tests, we identified the optimal combination of those physical options that produced the most realistic precipitation and temperature simulations for the CONUS. In addition to the physics sensitivity tests, two sets of 10-year (1949 - 1958) simulations for the CONUS were preformed. The first set was produced through continuous runs, while the second set was produced with 10 individual year runs. Each individual year includes a 16-month period from September of each year through December of the next year, and WRF was initialized on September 1 for these 10 separate runs. The first four-months of (September -December) model output accounts for model spin-up and is therefore discarded, while the simulations for the subsequent twelve months (January - December) were carried on for analysis. By comparing these two sets of simulations utilizing the above two different approaches, unrealistic climate drifts that are often generated from long-term, continuous model integration are effectively identified. A statistical comparison between both approaches and observed historical values is also presented in this study.
Evaluation of the Weather Research and Forecasting Model over the Contingent United Sates
Eccles Conference Center
This study is intended to examine the performance of the next-generation Weather Research and Forecasting (WRF) model version 3.2 over the Contingent United States (CONUS). WRF was developed by the National Center for Atmospheric Research and has been broadly used for weather, climate, and water resources research and prediction. WRF's ability to combine user specified combinations of various physics and model settings is the main reason it is the preferred model by climate scientists, as it allows for a level of customization. Numerous permutations of the physics options within WRF were tested for the period of 1998 - 1999.
Through these tests, we identified the optimal combination of those physical options that produced the most realistic precipitation and temperature simulations for the CONUS. In addition to the physics sensitivity tests, two sets of 10-year (1949 - 1958) simulations for the CONUS were preformed. The first set was produced through continuous runs, while the second set was produced with 10 individual year runs. Each individual year includes a 16-month period from September of each year through December of the next year, and WRF was initialized on September 1 for these 10 separate runs. The first four-months of (September -December) model output accounts for model spin-up and is therefore discarded, while the simulations for the subsequent twelve months (January - December) were carried on for analysis. By comparing these two sets of simulations utilizing the above two different approaches, unrealistic climate drifts that are often generated from long-term, continuous model integration are effectively identified. A statistical comparison between both approaches and observed historical values is also presented in this study.
https://digitalcommons.usu.edu/runoff/2011/Posters/9